I think this is a bad idea because I’ve thought of something not explicitly mentioned in the article and can only assume my phone will lock me out with no way of ever being able to log in again!
I have no faith that they might consider outlying cases or that there are people it could serve well who don’t care about them.
Obviously a terrible idea even though I know next to nothing about how it will work in actual practice.
What could possibly go wrong?
So we replace passwords with giving the evil ones another great reason to spy on us? Yeah, how about, fuck that?
Reddit: “Please be a dick so we can confirm it’s really you.”
Don’t kid yourself. For the most part, it still is.
When you hear about Google’s cars self-driving millions of miles, it’s almost entirely in the same small area - an area with all the objects mapped out, including the overhead locations of all the traffic lights so the car knows where to look for them. And lots of wireless connectivity for the cloud server back end. And driving only in good conditions, no rain.
There’s much it can’t handle, like construction zones or police office hand-signalling to pull over, and even merging onto a busy freeway is a problem.
Here in Canada we have snow half the year. Dividing lines on the roads disappear, the edges of the roads are vague at best, and snow banks along the roads can change on a daily basis, contradicting the 3D object map data. Connectivity is far from guaranteed.
Yes, they’re making great progress. In the real world, but only under optimum conditions. Out in the dirt road countryside and in winter, it’s largely still crazy talk.
I did read a couple months ago about one auto manufacturer experimenting with self-driving in winter conditions. The car would rely much more on the 3D object map - the pre-mapped locations of large objects along the road.
Which brings us to the debate over who gets sued in a self-driving car accident: The passenger, the vehicle owner, the taxi service, the car manufacturer or the software developer. I’m waiting for the other option: The property owner who moves their shed a couple feet, not quite contradicting the car’s 3D-object map cloud service, but instead leading the car to compensate by moving over a couple feet and hitting something.
I’d also like to see what happens in traffic with LOTS of self-driving cars - the similar RADAR and LIDAR from 50 cars all bouncing back signals at each other.
> Hate passwords?
Well, no. Seriously. They do what I need, and I’m happy with my password management.
> The trust score is based off various user-specific data points, including current location, facial recognition, and typing patterns.
I have location turned off unless I’m actively using a map, never use a device’s camera - never mind for facial recognition - and my usage pattern means I rarely have a reason to type much.
To be honest, I make a point of avoiding anything that ties me to a particular device - being able to log into a browser-based service, rather than anything installed locally, is far more important to me.
I almost walked away from Twitter because they were demanding a phone number to “verify” I was who I said I was. If they don’t know who I am how the fuck do they know what my phone number is?
And for the record I don’t hate passwords. I hate things like Windows 10 recording information that should be strictly private and sending it places it doesn’t need to do.
No thank you, this will only increase the value of our personal data, which as it is has been collected by everyone under the sun. Recreating someone’s data and usage patterns on another device would be relatively easy.
It sounds like a fine idea provided it’s not mandatory and that failing identification you can provide a backup password…and don’t mind Google looking through your camera whenever it wants to.
Well thought i’d just coin the term “baysean cock up”
What’s that?
Choose examples from almost any of the posts above mine.
Edited to add: Note, I’m not saying it’s a GOOD idea, just that while there are many reasons for particular people to not want to use it, I don’t see enough info to declare it a bad idea for everyone all around.
Some people say fingerprint ID on my phone is a terrible idea from the get go, but for my usage, it’s exactly want I want. If I run into problems because of it, it’s on me and I’ll take the responsibility.
I just got a new phone. I went out of my way for a device without a fingerprint scanner because simply, what if I had a bandage on my finger because I was clumsy and couldn’t log into my phone. Biometrics are shiny and sexy right now, but a shitty way to lock out your phone in case you need to let someone else use it or whatever.
So yeah, I’m not behind this business at all.
I would need bandages on five separate fingers for it to lock me out. Which could happen.
At that point I can type in my password. Assuming my fingers are still working after all the bandages. =o)
Fair enough. As an option only, the trust score mechanic might play out okay. For my own experience, there are too many potential pitfalls for trust scores to be the only access point for any device of mine. But YMMV. 
Mine has one and I refuse to use it; opting instead for the 6 digit code.
Color me paranoid, but I’m just not okay with Google or any public corporation having a record of my fingerprint… Plus the really practical reason you stated above.
I’ve had my hand bandaged from a burn before and had to work things with my other hand. All in all biometrics are more security theatre than secure. When things go awry are you going to get new fingers as easily as I replace a password?
You can go into settings and bypass the fingerprint scanner entirely. That’s what I do, because my fingerprints are too shallow to read consistently.
Your fingerprint has expired and must be changed.
Your fingerprint must be at least seven digits and cannot repeat your previous fingerprint; must contain arches, loops, or whorls, and cannot contain your full hand …
Should reduce drunk texting!
